pranaydeeps/lettuce_pos_nl_mono

pos_final_mono_nl

This model is a fine-tuned version of pdelobelle/robbert-v2-dutch-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1115
• Precision: 0.9783
• Recall: 0.9784
• F1: 0.9783
• Accuracy: 0.9791

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 256
• eval_batch_size: 256
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 1024
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 40.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	69	3.7703	0.2597	0.1252	0.1689	0.2575
No log	2.0	138	1.0148	0.8058	0.8008	0.8033	0.8066
No log	3.0	207	0.3402	0.9302	0.9278	0.9290	0.9299
No log	4.0	276	0.2016	0.9559	0.9551	0.9555	0.9561
No log	5.0	345	0.1486	0.9643	0.9638	0.9641	0.9648
No log	6.0	414	0.1206	0.9697	0.9696	0.9697	0.9702
No log	7.0	483	0.1063	0.9720	0.9719	0.9720	0.9727
1.2192	8.0	552	0.0983	0.9734	0.9735	0.9735	0.9742
1.2192	9.0	621	0.0947	0.9746	0.9747	0.9746	0.9754
1.2192	10.0	690	0.0913	0.9753	0.9755	0.9754	0.9761
1.2192	11.0	759	0.0885	0.9761	0.9763	0.9762	0.9770
1.2192	12.0	828	0.0877	0.9764	0.9765	0.9764	0.9772
1.2192	13.0	897	0.0878	0.9767	0.9769	0.9768	0.9775
1.2192	14.0	966	0.0873	0.9767	0.9769	0.9768	0.9776
0.0688	15.0	1035	0.0877	0.9771	0.9773	0.9772	0.9779
0.0688	16.0	1104	0.0878	0.9773	0.9774	0.9773	0.9781
0.0688	17.0	1173	0.0897	0.9772	0.9773	0.9773	0.9781
0.0688	18.0	1242	0.0909	0.9775	0.9776	0.9776	0.9783
0.0688	19.0	1311	0.0917	0.9776	0.9778	0.9777	0.9785
0.0688	20.0	1380	0.0924	0.9778	0.9780	0.9779	0.9787
0.0688	21.0	1449	0.0949	0.9777	0.9779	0.9778	0.9785
0.0366	22.0	1518	0.0956	0.9776	0.9777	0.9777	0.9784
0.0366	23.0	1587	0.0962	0.9778	0.9780	0.9779	0.9786
0.0366	24.0	1656	0.0992	0.9777	0.9780	0.9779	0.9786
0.0366	25.0	1725	0.0999	0.9779	0.9781	0.9780	0.9787
0.0366	26.0	1794	0.1007	0.9780	0.9782	0.9781	0.9789
0.0366	27.0	1863	0.1022	0.9781	0.9782	0.9782	0.9789
0.0366	28.0	1932	0.1030	0.9781	0.9783	0.9782	0.9790
0.0226	29.0	2001	0.1055	0.9781	0.9782	0.9781	0.9789
0.0226	30.0	2070	0.1057	0.9780	0.9782	0.9781	0.9789
0.0226	31.0	2139	0.1067	0.9780	0.9781	0.9780	0.9788
0.0226	32.0	2208	0.1077	0.9780	0.9782	0.9781	0.9789
0.0226	33.0	2277	0.1085	0.9780	0.9781	0.9781	0.9789
0.0226	34.0	2346	0.1094	0.9781	0.9782	0.9781	0.9789
0.0226	35.0	2415	0.1095	0.9783	0.9784	0.9783	0.9791
0.0226	36.0	2484	0.1101	0.9780	0.9782	0.9781	0.9789
0.0159	37.0	2553	0.1114	0.9782	0.9784	0.9783	0.9791
0.0159	38.0	2622	0.1111	0.9782	0.9784	0.9783	0.9791
0.0159	39.0	2691	0.1114	0.9782	0.9784	0.9783	0.9791
0.0159	40.0	2760	0.1115	0.9783	0.9784	0.9783	0.9791

Framework versions

• Transformers 4.25.1
• Pytorch 1.12.0
• Datasets 2.18.0
• Tokenizers 0.13.2